A project onboarding command for a React Native development agent. It checks required software and then adds the project-side configuration the agent needs.

Use it when setting up a new React Native project, checking iOS or Android prerequisites, and configuring device-control support.

It finds setup problems early and handles required wiring so the plugin can work without manually following its documentation.

setup — Command by Lykhoyda

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/lykhoyda/rn-dev-agent/setup

Clone the repo

git clone --depth 1 https://github.com/Lykhoyda/rn-dev-agent

Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 5,861 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00041	$0.05861
Opus 5	$0.00020	$0.02930
Sonnet 5	$0.00008	$0.01172
Haiku 4.5	$0.00004	$0.00586

Measured yesterday against content hash 41229e5ae89d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

packages/claude-plugin/commands/setup.md · 231 lines

How it starts

The opening of the file, as written. The whole thing — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Run the rn-dev-agent onboarding flow. Two phases: diagnose what's installed, then inject the project-side wiring the plugin needs. Most users don't read the README — this command does the wiring for them.

Phase 1 — Diagnose

Invoke the rn-setup skill (same as /rn-dev-agent:doctor). Walk every prerequisite check, starting at check 0, and present the table.

Abort thresholds. Only the rows that block ALL plugin functionality count as critical for onboarding:

CRITICAL (abort if any fail): source declaration for a non-Git app root (check 0 — name missing RN_DEV_AGENT_DECLARED_ROOT or RN_DEV_AGENT_DECLARED_MANIFESTS declarations and point to the session-authority contract before aborting injection or build), Node.js version, CDP bridge dependencies. Platform-specific device-control row: rn-fast-runner build artifacts on macOS targeting iOS, OR rn-android-runner build/install when targeting Android. Both runners ship in-tree — there is no external CLI to install; they build/install on first use. iOS-only users don't need rn-android-runner; Android-only users don't need rn-fast-runner. Without the relevant device-control row green, the plugin can't drive the simulator/emulator. Show the one-time pre-build commands (if any) and ask the user to run them, then re-run /rn-dev-agent:setup.
DEFERRED (warn but continue): Metro dev server, iOS/Android simulator/emulator, CDP connection. A clean-clone user running /setup immediately after git clone typically hasn't started Metro or booted a simulator yet — that's a normal onboarding state, not a failure. The CLAUDE.md / nav-ref / Zustand / scaffold injection steps don't need Metro running. Phase 2 Step E verification (cdp_status) WILL skip with a note if Metro+simulator aren't up; the user can come back to it later.
OPTIONAL (note, continue): ffmpeg, physical-device prerequisites when no devices connected, the off-platform device row (e.g. rn-android-runner on an iOS-only macOS setup → N/A).

Read the full file on GitHub · 231 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

yesterday First seen · 231 lines · 41 tokens per session scan A 41229e5ae89d

Subscribe to this mod's changes

setup is a command published in the GitHub repository Lykhoyda/rn-dev-agent (11 stars, last pushed 2d ago), licensed MIT. It adds 41 tokens to every session and 5,861 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Debugging Frontend Testing agent agent-framework agent-orchestration agentic-ai agentic-workflow agents

setup

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Phase 1 — Diagnose

What this file has done since we first saw it

Other commands, from other repositories

dashboard-flow-auto

dashboard-flow-abort

dashboard-git-branches

brainstorm

discover

propose